Background: When older adults are hospitalized, they are often discharged on substantially different medication regimens than the ones they were using prior to admission. In this setting, errors of medication reconciliation and problems with medication adherence are common. To address these problems, much effort has focused on improving medication reconciliation and communication at the time of discharge. However, little attention has been paid to the medication changes themselves that happen during the hospital stay. These changes are often inappropriate, and can lead to medication confusion, non-adherence, and adverse drug reactions in the weeks and months after hospital discharge. This is particularly important for chronic diseases such as hypertension and diabetes, as medicines for these conditions may often be changed during the hospital stay and perpetuated at discharge despite having little relation to the reason for admission. Aims: (1) To determine the frequency and epidemiology of changes to medication regimens for hypertension and diabetes in older adults admitted to VA hospitals; (2) To evaluate the impact of these medication changes on medication persistence and adherence in the year after discharge; and (3) To evaluate the association between changes to chronic medications during the hospital stay and subsequent emergency department visits and hospital readmissions. Methods: Using national data from VA and Medicare, we will assemble a retrospective cohort of veterans age 65 years and older who had hypertension or diabetes and were hospitalized in a VA medical center with community-acquired pneumonia, urinary tract infection, or venous thromboembolism. (These conditions were chosen because they are common and are treated with a different set of medications than those used to manage hypertension and diabetes). Using claims data and targeted chart review, we will identify 500 patients who were discharged on more aggressive regimens for blood pressure and/or glycemic control than they were receiving prior to admission, and a control group of 500 patients without intensified regimens. We will use standard methods to evaluate persistence of and adherence to medications in the year after hospital discharge. We will use Cox regression and mixed effects generalized linear models to compare medication persistence and adherence in patients who were discharged from the hospital with a more intensive medication regimen vs. patients whose medication regimens were not intensified. We will use similar methods to determine the impact of medication changes during hospitalization on subsequent emergency department use and hospital readmission.